A Data-Quality Audit Before Migrating to EDGEBIC

A data-quality audit is a read-only assessment of your RMDB data that scores its readiness and lists every gap before you clean or import anything, so you know the true size of the migration up front. EDGEBIC by User Solutions schedules against finite capacity, which depends on complete routings, real times, and work centers with capacity, so the audit checks exactly those and flags what is missing. Do this before cleaning and before import. It converts "the migration will probably be fine" into a scored, prioritized findings list you can plan against.

Audit first, clean second

There is a natural instinct to jump straight into fixing data. Resist it for one step. The audit and the cleaning are two different jobs: the audit assesses and scores without changing anything, and the cleaning pass remediates what the audit found. Separating them matters because you cannot plan a cleanup you have not measured. The audit tells you whether you face two days or two weeks of work, and where to spend it.

Keep the audit read-only. You are inventorying and scoring, not editing. That discipline keeps the assessment honest and gives you a clean baseline to measure the cleanup against.

What the finite capacity engine actually needs

The audit is not a generic data-hygiene exercise. It checks the specific things that make or break a finite capacity schedule. Everything else is secondary.

Complete routings. Every product needs its full operation sequence, in order, with a work center named for each step. A routing missing a step schedules short; a routing missing a work center cannot schedule at all. This is the first thing to score.

Real times. Each operation needs a setup time and a run time per piece, and they need to be plausible. A blank time or an obviously wrong one produces a schedule that looks precise and is not. Times are the second score.

Work center capacity. Every work center needs its hours, its number of identical instances, and its utilization entered. A work center with no capacity is invisible to the engine.

Unique item numbers. Duplicate item numbers collide on import and cross-wire routings to the wrong product. Cheap to check, expensive to miss.

Score these four, and you have measured migration readiness where it counts.

How to audit at scale

Thousands of items cannot be checked by hand, so audit two ways at once.

By volume. Sort items by production volume and hand-check the top slice. Your busiest hundred or so products drive most of the schedule, and their data has to be right. Detailed verification here is worth the time.

By exception. Run filters across the whole set in the spreadsheet to surface specific defects: routings with no setup time, operations with no work center, duplicate item numbers, work centers with zero capacity. These filters find the blocking gaps without reading every row.

The two together give a scored picture: high-volume data verified in detail, and the full set screened for the exact defects the engine cannot tolerate.

Scoring the findings

Turn the checks into a simple scored table so the effort is visible and prioritized.

Rank the findings by severity. Blocking gaps come first, because no pilot runs until they are closed. Plausibility issues come second, because they produce quiet errors rather than hard failures. This ranked list is the audit's deliverable, and it feeds straight into the cleanup and the migration timeline.

Why RMDB tolerating it does not mean it is clean

A common objection: the data has scheduled in RMDB for years, so why audit it? Because a system can tolerate a gap a planner quietly filled in by habit. RMDB may have accepted a blank setup time because a scheduler always knew the number. EDGEBIC needs it as explicit data, since the engine cannot read a planner's memory. The move is the moment to make the implicit explicit, and the audit is how you find what was implicit before it surfaces as a strange date in the pilot.

What the audit is not

The audit does not clean, and it does not import. It also does not need to be exhaustive on low-volume, low-risk data: a rarely made product with a slightly stale time is a cleanup footnote, not an audit blocker. Keep the audit focused on what blocks or distorts scheduling, produce the scored list, and hand it to the cleanup. Over-auditing trivia delays the pilot as surely as skipping the audit does.

The takeaway

A data-quality audit before EDGEBIC is a read-only, scored assessment of your RMDB routings, times, work center capacity, and item numbers that tells you the true size of the migration before you clean or import. Audit by volume and by exception, rank the findings by severity, and hand the list to the cleanup. It is cheaper to find a missing setup time in an audit than in a pilot. See the platform on the EDGEBIC overview, read the upgrade path on the RMDB to EDGEBIC guide, and move from findings to fixes with cleaning your data before importing to EDGEBIC.